COLOU1% INHEP~ITANCE IN . II. THE PIE- BALD PATTEI~N OF THE PIEBALD BI%EED.

B~r J. A. FI~ASEtI I%OBEI~TS. F~'om the Depa~'tment of Ag~'ieultu~'e, Unive~'sity College of No~'th Wales, Bangor', and the A~gmaZ 2~'eediq~g _Reseaq'eh Depa~'tment, Unive'rsity of Edinbu~'gh. (With Two Pla~es.)

CONTENTS. PAGE 1. The Piebald Breed of Sheep 77 2. Crosses with O~her 2Breeds. The F 1 Generation 79 3. Further ExperimentM Breeding 80 4. Discussion 81 5. Summ~ry 82 P~eferencos 82 Description of Plates X and XI 82 1. The Piebald B,reed of Shee~). Tins aneienb ornamental breed has been variously described as Spanish, Syrian, Portuguese, African, Z!flu , Persian, Egyptian, Barbary, and.also as "Jacob's Sheep." Owing to the uncertainty regarding the origin, of the breed, it is perhaps best to follow Elwes (1913) in using the non- committal name "Piebald;" These sheep possess many excellent economic qualities and it is doubtless due to this fact as well as to their picturesque appearance that so many flocks are in existence in this country, Wallace (1925) putting the number of flocks as high as one hundred and fifty. Piebald sheep have been in existence in England for a considerable period, and pictures at Tabley House and at Wentworth painted about 1760 show that there has been little change in appearance since that time. Painstaking efforts to trace the breed in other countries have fated, and their country of origin must remain a matter of considerable doubt. There is some evidence of a North African origin, while the ease with which South African sheep imported last century were absorbed into certain flocks indicate the possibility of a relationship. A full account of his efforts to trace the history of the breed is given by Noble (1913); descriptions of the sheep will be found in the writings of Elwes (1913), Wallace (1923), Lydekker (1912), and Portal (1923), while 78 Oolour Znheritanee in Shee~) Plates X and XI give some idea of their appearance..Attention will be mMnly confined in the present paper to the piebMd pattern which is the outstanding breed characteristic. This patffern takes the form of black patches on a white grotmd, that is to say ffhe black patches are usually rounded, giving an effect as though the pigment tended ffo spread outwards from a number of points.. There is lit~,le tendency for definite arrangement to occm', except theft it is usual for ~,]:m black areas to involve the two sides of the face, leaving a clear st:rip of whiffs down the centre. The ewe shown in Plate XI, fig. 9, i s the most typicM in this respect. This arrangemen'% however, is not invariable. There is wide variation in the total amount of spoffting, as is Shown in Plafie X, figs. ]-3, and there is, perhaps, some tendency for the pigment, as it increases in total amount, to spread backwards from the forepart of the body towards ffhe hindquarters. The nature of the spotting is also variable, in that the black patches may be comparatively small, most of the sheep ilIustrated in the paper coming into ffhis category; or, on the other hand, the patches may be large, so that ffhe pattern consists of just a very few extremely large patches spreading over the whole body. An excellent example will be found in a~ illustration given by Etwes (1913) and reproduced by Wallace (1923). It is probable that the amount and nature of the spotting is readily amenable ~o selection and the varying appearance of different flocks is due to the type of pattern that has been chosen as the ideal. Piebald sheep breed remarkably true, the writer having quiffe failed to discover any instance of a lamb other than Piebald being born in a Piebald flock. One breeder writes "...The farm steward tells me that he has never seen one of these sheep wholly black or wholly white." This reply is typical. Such observations as the writer has been able go make appeared to show thaff the correlation between the ~ype of spotting in mother and offspring is not very high. The black patches in new-born lambs are as dark as the black exhibited by offher black sheep, but as the lamb grows ~he colonr becomes transformed infio a brown or even a fawn, this change being much more marked than the usual bleaching see~ in ordhlary black sheep. It might be mentioned that while the sheep in the majority of flocks are four-horned a number of flocks are in exisffence which consisff solely of ffwo-horned sheep, so that this feature cannot be regarded as a breed characteristic, and may possibly have even been introduced by crosses with other primitive fou>horned breeds. There is no doubt, however, that certain :[locks have been :[our-horned for a very long period. J. A. Fm~sJ~ Rom~I{TS 79

2. Urosses with Other B,reeds. The F:t Generation. It has been frequently recorded that ~he result of crosses with other breeds is the production oti self-blacks only, and that many of tile ~'l's exhibit "white pattern" which consists of a whibe spot on the Cop of the head and a white tip to the ~ail. Noble (1.913)states: "Now i'~ is a curious fact ~hab wl/cn our piebMd sheep a,re crossed with any other domestic breeds, t~he lambs a.re practically all born black with a white; patch on the forehead and partly white tails." Elwes (29].3) quotes this sta,~ement and mentions the fact that he crossed piebald rams to ewes of many bm;eds apparently wil~h I~he same resull). PortaI (19%) states tha.t self-blacks result :from crosses with [-[ighland sheep, , Southdown., and Cheviot slheep. I-[e aIso stages that two lambs item a wild Mouitlon rant and a Piebald ewe were self-blacks with white pattern. Mr I%. Holland Martin, of Overbury Court, Tewkesbury, informs me that his PiebMd sheep have been crossed with Oxford Downs in previous years with the same result. In the autumn of 1923, Major E. J. W. Plata, of Gorddinog, Elan- fairfschan, wishing to test this statement, made reciprocal crosses between Piebald sheep on the one hand, and Southdown and Welsh 3{ouubaiu sl~eep on the other. The writer had the opportunity of ex- amining the lambs: all were black, though in the e~se of the Welsh crosses one or awe showed very extensive white pattern. These lambs possessed several little white tufts especially about the shoulders. It is probable bhab the white tufts wend only exist in the lamb's coat. The writer also had the opportunity of seeing some Piebald-Kerryhill F,'s born in the flock of Sir C4srald Corbebt, Bart., Aeton l%synauld, Shrews- bury. Fifty of these lambs were raised, and all were black; in six eases only was the white pattern so extensive as to be specially noted. To ~hs general tale that the F[s of any cross are self-black, the writer knows of two exceptions. The first is recorded by Elwes (1913) who mentions one particular Piebald ram which when mated to Pat-rumped ewes gave four blacks and one white. The same ram mated to a Wiltshire x Sony ewe gave one piebald lamb and one white. The other case came under the writer's personal observation. Major Platf, in 192,l: crossed a Dorset ram go two Piebald ewes, the result being four piebald lambs. The same ram mated to another ewe in 1925 sired two more lambs, both piebald. These lambs are shown in Plate XI, figs. 8-11. Ig will be seen that though some are perhaps not typical, they fail within the limits of variation of the pattern. Oln_e additional point is worbl~y of mention. It has been noted above 80 Uolour ],nheritcsnee in Sheep that it is characteristic that the black eolour of the pigmented patches in the pure breed becomes transformed into a brown or fawn. This does not appear to be the ease in the black coat of the 2Fl's. It has only been possible to examine two F i Sheep that have attained the age of a year, and both ]possessed. coats at that time that were no lighter than those usually seen in black sheep. 3. Further Experimental Bree~gng. To obtain further information as to the inheritance of the piebald pattern, _Fi rams were back-crossed to white and to piebald sheep respectively. In the first ease a Piebald-Southdown/v 1 ram was crossed to white Welsh Mountain ewes, this experiment being carried out by Najor Platt. In the second ease a Piebald-Black Welsh Mountain _Fi ram i was crossed to Piebald ewes, this experiment being carried out at ~he farm of the University College of North Wales, Bangor. The results were as follows: TABLE I. Self-black F i c~ x White 9" Sex Pheno~ype Ewe of lamb of lamb White pattern 1 3 Wtfite 1 ~ White 2 c~ Black None 3 .9 Black White tuf~ on head only ~ White 5 3 White 5 3 Black WtKte tuf~ on head only 6 -9 Black White tung on head only 6 -9 Wlfite 7 -9 Black }Vhi~e ~uft on head only 7 9- White TABLE II. Self-black F i ~ x PiebMd -9. Sex Phenotype Ewe of lamb of lamb White pattern S 37 c~ Piebald -- S 38 -9 Black Head oMy S 39 c~ Black Head only S "39 -9 Piebald -- S 40 -9 Black Head only S ~1:0 -9 Black t[ead only S 41:l _9 Piebald -- S 4-1 ~ Piebald -- S ~2 c~ Piebald -- S 42 .9 Piebald -- i The choice of a Black Welsh x Piebald ram for the second back-cross rather than a Piebald x white ram was clue to the fact that this ram was also being crossed to white ewes to tesg whether the dominant black of.'the two breeds depends on the same factor, and in this way it was possible to economise space a~ tupping time by using the one ram for the two purposes. J. A. ]?IAsi~i~ Igo~m~,rs 81 In the case of the back-cross to white, segregation was perfect. The white pattern exhibited by the blacks was nob extensive and did not involve ~he tail in any case (Plate XI, figs. 6-7). The F 1 rain used in the cross did not exhibit the pattern at all. The four blacks resulting from the back-cross to piebald were ordinary blacks, but in this case bhc white pattern was more extensive (Plate X, fig. 5). It still did no~ involve the tail. Five of the six piebalds were well within the normal limits of the spotting, bug one lamb was cx:tremely darlq the whi~,e areas being reduced to a more or less white face and a few extremely small white areas on the body, which, however, involved mol'e than the lamb's coat, so th~ there is little doubt that this lamb was genelffcally piebald and not black. The F 1 ram in this case had a/So exhibited fairly extensive white patbern on the hea.d, the white patch including a small are:, of 8he sti:ff hairs of the face, so that tilts small area of white was persist, cut.

4-. Discussion. The numbers involved in the experiment are small, but it may be Stated with considerable eonficlenee that Piebald sheep differ from ordinary white sheep in that they possess a dominant black factor and a recessive pattern factor. The back-cross ratios, six whites be five blacks in the back-cross to white, and six piebalds to four blacks in the back-cross to piebald, are sufficiently good evidence of this. The two cases quoted above where Piebald Fi's appeared, are to be explained on the assumption that the white parent in each case possessed the pattern faot0r. This is confirmed by an experiment mentioned by Ewart (1919) Where some of the ~s sheep of a Southdown • Soay cross were piebald. I~ is significant that in the ease of Elwss' piebald F1, the mother was a crossbred Soay. The experiments alone do not directly answer the question as to whether a white sheep homozygous for the pa,ttern factor wotdd still be white, but this is very probable. The Dorset t-Iorn ram ~ sired six piebalds would most probably be homozygous for the pattern factor, and the fact that piebalds do not occur in ordinary white flocks strengthens ~his argument. It may be accepted then as reasonably certain.~shag the piebald pattern factor o~fly produces an effect on a sheep tk~g is already pigmented. It is probable that the extreme range of variation in the extent and nature of the spotting is to be ascribed to a number of modifying factors and that a certain amount of variation may be non-genetic (Sewall Wright). The constancy with ~

case of 5he recessive pat~ez'n factor but not so in the case of ~he black factor. The case of Elwes' ram which sired two white lambs is to be explained on the assumption that he was heterozygons for the black factor. The two Piebald- F~ lambs born in 1.926 exhibited almost precisely similar markings (Plate XI, figs. 10-11). These markings, as far as the ventral sm'face is concerned, are not unlike those described by the writer and Jenkin (I926) and termed "reversed b:~dgerfaee" pattern. The writer's opinion is l;hat this resemblance is accidental or might be the expression of a general tendency go exhibit a light-coloured belly, and this opinion is strengthened by the fact that t]le fern' lambs sired by the same ram and beret in 1925 were in :He special[ way re~narkable as regards their spotting. White pattern is of interest because of its universal distribution in the case of coloured sheep. It is most fully dealt with by Adametz (1917) who considers that it depends upon a single recessive factor together with another main modifying factor (see also Elwes, 1912, 1913 ; Wallace, 1915; and l%oberts, 192@ The only special point in the present work is the frequency with which the head and tail markings are not associated. It is hoped to deal with the question of the relation of the dominant black of the Piebald breed ~o that of the Black Welsh breed at another time, and it is also hoped that data will then be presented which will indicate whether the bleaching of the black of the Piebald sheep is capable of a simple genetic explanation. The writer wishes to acknowledge the help received at all times from Prof. 1%. G. W~nite, of the University College of North Wales. I-Is is also indebted to Major Platt and to Major Platt's farm manager, Mr Coward, for [he facilities for observation placed a~ his disposal, for the information they have s/~pplied and for so readily consenting to carry the experiment further. The writer is indebted to iVI~ Holland Martin and to Mr Hall, of College Hill, Shrewsbary, for much useftd information.

1. A brief description is given of the Piebald breed of sheep. While there is a considerable range of variation in the extent and nature of the spotting these sheep breed remarkably true for the piebald character. 2. The. result of crosses with other breeds is the production of self- blacks. 3. A back-cross of the Fa to whigs gave six whites and five blacks. A back-cross of the F~ to piebald gave six piebalds and four blacks. J. :~. I~I~XSER ' I:~OBEI~TS 83

4. Piebald sheep differ, from Other breeds in that they possess a dominant black factor and a ~ecessive pattern factor which restricts the black to certai~/rent.

I%EFEI~ENCES. AD&m]:rz, .L. (1917). "Studien tiber die ig[endelsehe Vererbung der wiehgigsfien l~assenmerkm~le tier t~:~'akulschaf bet l%einzucht und Kreuzungen mig l~am- b0rfillets." Bibliot. Ge'uet. Vol. z. pp. 1-258. Er~w~s, .IT.. 3. (]912). "lgoi;es on the Primitive ~reeds of Sheep in Scot,land." Scot. Nat~o'. Jan.-Mareh, pp. 1--52. (t913). 6bride to the Primitive Breeds of Sheep and their U~'osses on ?g'xhibition ~t the Roy. Age"to. ~be. Show, ]J.risml. Edinburgh. Ewes% J. C. (1919). "The :[ntercrossh~g of Sheep ~md the Evolution of New Varieties of Wool." Scot. Jou~'n. Ag~'ie. Vol. ~r. LYDEIrA~ER, i%. (1912). The She@ and its 6'ousins, pp. 1-3/5. London. NO~L~, H. (1913). An Effort to Tmee the History of the so.called "Piebald" Spa~zish Sheep. Published privately i. :PoR~o M. (1923). "The Spanish or Spot~ed Sheep." Country Life, Vol. L~. Sept. p. 15. I~O~RTS, J. A. !~. (192~). "Colour Inheritance in Sheep. I. ])lack Colour ~md Badger-face P~ttel'n in Welsh ?/Ionntain Sheep." Journ. 6%neties, Vol. XlV. pp. 367-374. I~omm'~s, J. A. F. and J~g_~, T. J. (1926). "A Preliminary Note on Reversed ~adger-faee Pattern in Sheep." Welsh Jo~rn. Attic. Vol. l~. WA~0~ , 1%. (t915). "P~esults of Crossing Karakul Sheep with :European and Especially British Breeds." Jo,~rn. Boa~'d Agric. Eng. Vol. xx~. pp. ~3~-477. ~ :(!923)) 2arm Live Stock of U~'eat t~r:itain. OHver and Boyd.

DESCRIPTION OF PLATES X AND XI.

PL&TE X. F~g] 1. Lightest Piebald tamb in l~IQor Platb's flock, 1926. Fig. 2. Darkest Piebald lamb in ~{a~or Ph%tt's flock, 1926. F{g, 3. Piebald ewe with Iambs showing medium spottfllg. Figii ( Back-cross. Black F 1 xPiebald--piebald lamb. ~i~i 5. B~ok-c~os~. ~-1~ck ~ • Pi~b,~id--blaok l~mh. ~te~s~e wl~e pattorn o~ head.

PLAT~ XI. Fi~ 6/i Welsh ~IountMn- ewe ~dth l~mbs from Piebald-Southdown ~1 ram. Figi 7 i Black lamb (Black F 1 • wlfite back-cross) showing restricted wlfi~e pa~ern on head. Figsi 8~9; Piebald ewes wi~h four piebald 2~z lambs aired by Dorset Horn ram, 1925. FigS~ I0:=ll, F 1 piebald lamb sired by same Dorset Horn ram, t926.

This pamphlet has been duplicated by the Animal Breeding l%eseareh Department, 1926; Copies ~re av~ilable. JOURNAL OF 6ENETIO8, VOL. XVII. NO. 1. PLATE X JOURNAL OF GENETI08, VOL. XVll. NO. 1. PLATE X~